{"id":50276006,"url":"https://github.com/sbdk-dev/sbdk.dev","last_synced_at":"2026-05-27T20:32:21.722Z","repository":{"id":325978078,"uuid":"1083394394","full_name":"sbdk-dev/sbdk.dev","owner":"sbdk-dev","description":"A complete reference implementation of a local-first ecosystem for AI-powered analytics. This repository contains the source code for the SBDK.dev website, the central hub for the SBDK suite of open-source tools.","archived":false,"fork":false,"pushed_at":"2025-11-25T01:58:50.000Z","size":326,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-11-28T04:37:26.933Z","etag":null,"topics":["ai-powered-analytics","data","data-engineering","data-engineeringlocal-first","data-pipeline-automation","data-pipelines","dbt","dlt","duckdb","elt","etl-pipeline","llm","local-first","machine-learning","pipeline","sbdk","semantic-layer"],"latest_commit_sha":null,"homepage":"https://www.sbdk.dev","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/sbdk-dev.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-10-25T23:21:26.000Z","updated_at":"2025-11-25T04:11:13.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/sbdk-dev/sbdk.dev","commit_stats":null,"previous_names":["sbdk-dev/sbdk.dev"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/sbdk-dev/sbdk.dev","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sbdk-dev%2Fsbdk.dev","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sbdk-dev%2Fsbdk.dev/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sbdk-dev%2Fsbdk.dev/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sbdk-dev%2Fsbdk.dev/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sbdk-dev","download_url":"https://codeload.github.com/sbdk-dev/sbdk.dev/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sbdk-dev%2Fsbdk.dev/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33583394,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-05-27T02:00:06.184Z","response_time":53,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-powered-analytics","data","data-engineering","data-engineeringlocal-first","data-pipeline-automation","data-pipelines","dbt","dlt","duckdb","elt","etl-pipeline","llm","local-first","machine-learning","pipeline","sbdk","semantic-layer"],"created_at":"2026-05-27T20:32:20.773Z","updated_at":"2026-05-27T20:32:21.714Z","avatar_url":"https://github.com/sbdk-dev.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# SBDK.dev - Local-First Data \u0026 AI Reference Implementations\n\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n[![GitHub stars](https://img.shields.io/github/stars/sbdk-dev/sbdk-dev.svg?style=social\u0026label=Star)](https://github.com/sbdk-dev/sbdk-dev)\n\n**A complete open-source ecosystem demonstrating how to build local-first data and AI tools**\n\nThis repository serves as both the source code for [sbdk.dev](https://sbdk.dev) and a central hub for exploring five interconnected reference implementations that show how to build a complete local-first analytics platform—from data ingestion to AI-powered insights—all running on your laptop, without cloud dependencies.\n\n---\n\n## How The Ecosystem Fits Together\n\nEach project builds on the foundation to create a complete local-first analytics platform:\n\n```\n┌─────────────────────────────────────────────────────────────┐\n│  5. knowDB - MCP Integration                                │\n│     Connect to AI Assistants (Claude, ChatGPT)              │\n└─────────────────────────────────────────────────────────────┘\n                              ↑\n┌─────────────────────────────────────────────────────────────┐\n│  4. Local AI Analyst - Conversational Analytics             │\n│     Natural Language → Statistical Insights                 │\n└─────────────────────────────────────────────────────────────┘\n                              ↑\n┌─────────────────────────────────────────────────────────────┐\n│  3. Semantic Tracer - Visualization                         │\n│     Interactive Lineage Graphs for dbt Models               │\n└─────────────────────────────────────────────────────────────┘\n                              ↑\n┌─────────────────────────────────────────────────────────────┐\n│  2. Mallard (local-inference) - Intelligence                │\n│     ML/AI in SQL (Predictions, Embeddings, Explainability)  │\n└─────────────────────────────────────────────────────────────┘\n                              ↑\n┌─────────────────────────────────────────────────────────────┐\n│  1. SBDK.dev - Foundation                                   │\n│     Data Pipelines (DLT + dbt + DuckDB)                     │\n└─────────────────────────────────────────────────────────────┘\n```\n\n## The Projects\n\n### 1. 🏗️ SBDK.dev - The Foundation\n\n**Repository**: [sbdk-dev/sbdk-dev](https://github.com/sbdk-dev/sbdk-dev) | **Status**: Active\n\nThe core framework providing local-first data pipelines with DLT (ingestion), dbt (transformation), and DuckDB (analytics). Everything else builds on this foundation.\n\n**Key Features**:\n- **Lightning-Fast Setup**: Install and run in seconds with `uv`\n- **100% Local**: No cloud dependencies or complex configuration\n- **Complete Pipeline**: Ingestion → Transformation → Analytics in one toolkit\n- **Hot Reload**: Automatic re-runs when files change for iterative development\n\n**Use Case**: Start here if you're building data pipelines that need to run locally or learning the modern data stack (DLT, dbt, DuckDB).\n\n### 2. 🧠 Mallard (local-inference) - Intelligence Layer\n\n**Repository**: [sbdk-dev/local-inference](https://github.com/sbdk-dev/local-inference) | **Status**: Archived\n\nA DuckDB extension adding ML/AI capabilities directly in SQL. Run zero-shot predictions, generate embeddings, and get feature importance—all without separate ML infrastructure.\n\n**Key Features**:\n- **Zero-Shot ML**: Classification and regression without training\n- **SQL Interface**: All functionality exposed as SQL UDFs\n- **Rust Performance**: Built as a high-performance DuckDB extension\n- **Embeddings \u0026 Explainability**: Dense vectors and feature importance\n\n**Use Case**: Add ML capabilities to your data pipelines without complex ML infrastructure. Perfect for prototyping ML features or building \"Snowflake Cortex\" style analytics locally.\n\n### 3. 🔍 Semantic Tracer - Visualization\n\n**Repository**: [sbdk-dev/semantic-tracer](https://github.com/sbdk-dev/semantic-tracer) | **Status**: Archived\n\nVisualizes dbt semantic layers with interactive lineage graphs. Understand how your metrics, dimensions, and entities connect—all processed locally.\n\n**Key Features**:\n- **Interactive Graphs**: React Flow-based visualization of semantic models\n- **dbt Integration**: Direct connection to `semantic_models.yml`\n- **Tauri Desktop App**: Lightweight Rust backend with web frontend\n- **100% Local**: Semantic models and data never leave your machine\n\n**Use Case**: Understand complex dbt projects, document semantic relationships, or build similar visualization tools for data platforms.\n\n### 4. 💬 Local AI Analyst - Conversational Analytics\n\n**Repository**: [sbdk-dev/local-ai-analyst](https://github.com/sbdk-dev/local-ai-analyst) | **Status**: Archived\n\nAI-powered data analyst with statistical rigor. Ask questions in natural language, get answers based on real query results with confidence intervals and significance testing.\n\n**Key Features**:\n- **Natural Language Queries**: \"What's our conversion rate by plan type?\"\n- **Statistical Rigor**: Automatic significance testing and confidence intervals\n- **Execution-First**: Prevents AI hallucination by running queries first\n- **Multi-Query Workflows**: Complex analysis with multiple related queries\n\n**Use Case**: Build conversational analytics tools that prevent AI fabrication through statistical validation and execution-first approaches.\n\n### 5. 🔌 knowDB - Integration Layer\n\n**Repository**: [sbdk-dev/knowDB](https://github.com/sbdk-dev/knowDB) | **Status**: Archived\n\nConnects everything to AI assistants via MCP (Model Context Protocol). Query your data through Claude Desktop or ChatGPT Desktop with automatic dbt model syncing.\n\n**Key Features**:\n- **MCP Integration**: Works with Claude Desktop, ChatGPT Desktop, and any MCP client\n- **dbt Auto-Sync**: Automatic semantic layer synchronization\n- **Natural Language Queries**: Ask questions through your AI assistant\n- **Full Local**: All processing happens on your machine\n\n**Use Case**: Learn how to build MCP servers that connect data platforms to AI assistants, or fork to add MCP support to your own tools.\n\n---\n\n## Getting Started\n\n### Option 1: Start with the Foundation\n\n```bash\n# Clone and install SBDK.dev\ngit clone https://github.com/sbdk-dev/sbdk-dev\ncd sbdk-dev\npip install -e .\nsbdk init my-project\n```\n\n### Option 2: Fork Any Project\n\nEach project works standalone. Pick the one that matches your needs:\n\n1. **Building data pipelines?** → Start with [SBDK.dev](https://github.com/sbdk-dev/sbdk-dev)\n2. **Adding ML to your database?** → Check out [Mallard](https://github.com/sbdk-dev/local-inference)\n3. **Visualizing dbt models?** → Explore [Semantic Tracer](https://github.com/sbdk-dev/semantic-tracer)\n4. **Building conversational analytics?** → Fork [Local AI Analyst](https://github.com/sbdk-dev/local-ai-analyst)\n5. **Integrating with AI assistants?** → Try [knowDB](https://github.com/sbdk-dev/knowDB)\n\n### Option 3: Use as Learning Material\n\nAll projects include:\n- Complete, production-quality code\n- Comprehensive documentation\n- Real-world patterns and best practices\n- Examples and test cases\n\nPerfect for learning modern data engineering, Rust, MCP integration, or building local-first tools.\n\n---\n\n## Why These Projects Were Archived\n\nThese projects represent complete, proven implementations of local-first data tools. They're archived as reference implementations because:\n\n- **They're Complete**: Each project is production-quality and fully functional\n- **They Demonstrate Patterns**: Best practices for local-first, data engineering, and AI integration\n- **They're Ready to Fork**: Stable codebases perfect for adaptation and extension\n- **They Work Together**: Designed as an ecosystem but each works independently\n\nThe goal is to help others build similar tools, not to maintain these specific implementations indefinitely.\n\n---\n\n## Contributing\n\nWhile the individual projects are archived, we welcome:\n\n- **Bug reports and fixes** for critical issues\n- **Documentation improvements** to help others understand the code\n- **Showcase your fork**: Open an issue to share what you've built\n\nFor new features, please fork the project and build it yourself—that's what these are for!\n\n---\n\n## License\n\nAll projects in the SBDK ecosystem are MIT licensed. Use them however you want, commercially or personally, with or without attribution.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsbdk-dev%2Fsbdk.dev","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsbdk-dev%2Fsbdk.dev","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsbdk-dev%2Fsbdk.dev/lists"}